INDEX

Explanations

information related to technology and product releases

oai_token-act-pair · gpt-3.5-turbo

expressions of opportunity and potential

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 progresses

-0.63

 unforeseen

-0.61

ulative

-0.59

 interactions

-0.58

 manipulated

-0.55

 evolves

-0.54

 semblance

-0.54

lasting

-0.54

 perceptions

-0.53

verages

-0.53

POSITIVE LOGITS

 probably

0.85

starter

0.83

 surely

0.78

 largeDownload

0.77

sure

0.76

 suffice

0.72

probably

0.72

 definitely

0.71

 certainly

0.70

Probably

0.69

Activations Density 0.292%

information related to technology and product releases

expressions of opportunity and potential

No Comments

No Known Activations